Boundary layer — Prandtl's concept, growth along flat plate: Visual walkthrough

2.2.20 · D2 · Physics › Fluid Mechanics › Boundary layer — Prandtl's concept, growth along flat plate

Step 1 — The plate, the stream, and the two speeds

KYA. Ek flat plate ko flat rakha hua draw karo, jiske upar fluid left-to-right ek steady speed se beh raha hai. Uss door wali speed ko $U$ kaho (units: metres per second, m/s). Ab plate ko touch karne wale fluid mein zoom karo.

KYUN. Kisi bhi maths se pehle, hume ek single physical fact fix karni hai jo boundary layer ko exist karaati hai: wall se chipka hua fluid slide nahi kar sakta. Yahi no-slip condition hai — solid ke saath lagi fluid molecules ki layer ka velocity solid jaisa hi hota hai, aur stationary plate ke liye woh zero hota hai.

PICTURE. Figure mein, plate ke bahut upar wale amber arrows sab ek hi length ke hain ( $= U$ ). Plate ke bilkul upar wala arrow zero length ka hai. Toh humein do speeds bahut kareeb mili hain: $0$ wall par, $U$ upar.

Figure — Boundary layer — Prandtl's concept, growth along flat plate

Yahan $y$ ka matlab hai plate surface se seedha upar measured height; $y = 0$ khud wall hai.

Step 2 — The velocity profile: a squeeze from 0 to $U$

KYA. Ek fixed downstream position par $u$ ko height $y$ ke against plot karo. Yeh wall par $0$ se shuru hota hai aur $U$ tak bend karta hai.

KYUN. Hume dekhna hai ki $0$ se $U$ tak ka change kahan hota hai. Yeh har jagah nahi hota — yeh ek thin band ke across hota hai. Uss band ka vertical size hi woh quantity hai jise hum dhundh rahe hain.

PICTURE. Cyan curve wall ke paas steeply rise karta hai, phir $U$ ke paas jaate jaate flat ho jaata hai. Woh height mark karo jahan curve essentially $U$ tak pahunch jaata hai — woh height $δ$ hai.

Woh cyan curve ka wall ke paas steepness ek rate of change hai: $u$ har unit rise $y$ mein kitni speed gain karta hai. Hum ise $\frac{\partial u}{\partial y}$ likhte hain.

Step 3 — Why viscosity refuses to be ignored here

KYA. Uss steep gradient se ek physical force attach karo. Alag alag speed se move kar rahe neighbouring fluid layers ek doosre par rub karte hain; woh rubbing stress $τ = μ \frac{\partial u}{\partial y}$ hai (Newton's law of viscosity).

KYUN — tool choice. $μ$ ko introduce karna kyun zaroori hai jab air ka $μ$ microscopically small hai? Kyunki stress $μ$ times gradient hai, sirf $μ$ akela nahi. Step 2 mein humne dekha ki wall ke paas gradient $\sim U / δ$ hai, aur $δ$ tiny hai, toh gradient enormous hai. Ek microscopic $μ$ ko enormous gradient se multiply karo toh ek stress milta hai jo negligible nahi hai. Yahi d'Alembert's paradox ka resolution hai — drag is thin band mein rehta hai, outer flow mein nahi.

PICTURE. Fluid ki do sliding sheets; amber shear arrows opposite directions mein point karte hain; label small $μ$ dikhata hai lekin fat gradient se ek real force produce hoti hai.

Step 4 — Repackage viscosity as diffusion: meet $ν$

KYA. Viscosity $μ$ ko fluid ki density $ρ$ (mass per volume, kg/m³) se divide karo. Result ko kinematic viscosity $ν = μ / ρ$ kaho.

KYUN — tool choice. Hum abhi ek diffusion picture (Step 5) se argue karne wale hain, aur diffusion ko ek coefficient chahiye jo area-per-time (m²/s) mein measure ho, Pa·s mein nahi. Units check karo: $ν = \frac{μ}{ρ} = \frac{Pa\cdots}{kg/m ^{3}} = \frac{kg / ( m\cdots )}{kg / m ^{3}} = \frac{m ^{2}}{s}$ Yeh exactly "area per time" nikalta hai — ek spreading process ka signature. Toh $ν$ natural currency hai yeh samajhne ke liye ki "wall ka slow-down upar kitni fast spread karta hai."

PICTURE. Ek units-cancellation ledger: $μ$ (thick), $ρ$ (ek box mein mass ka stack), aur resulting $ν$ tagged m²/s, saath mein ek small dye blob spreading — diffusion coefficient ka visual meaning.

Step 5 — How far can slowing-down spread in the time available?

KYA. Do questions, combine kiye. (a) Ek fluid parcel plate length $x$ ko speed $\sim U$ se travel karta hai, toh woh wall ke saath $t \sim x / U$ time ke liye expose hota hai. (b) Us time mein, momentum-diffusion (coefficient $ν$ ) wall ke influence ko ek height $δ \sim ν t$ tak spread karta hai.

KYUN — tool choice. $ν t$ kyun aur $ν t$ kyun nahi? Kyunki koi bhi diffusion square root of time ki tarah spread karta hai: ek dye blob ka radius $D t$ ki tarah grow karta hai, heat $α t$ ki tarah creep karti hai, aur yahan momentum $ν t$ ki tarah. Diffusion ek random-walk spreading hai — spread double karne mein chaar guna time lagta hai. Yahi ek fact poori wajah hai ki $δ$ eventually $\propto x$ nikalta hai, kabhi linear nahi.

PICTURE. Upar: stream pe ride karta ek parcel, ek clock jisme $t = x / U$ dikh raha hai. Neeche: shaded slowed-region $ν t$ ki tarah upar fan kar raha hai, uski ceiling $δ$ hai.

Step 6 — Combine into the growth law

KYA. $t = x / U$ ko $δ \sim ν t$ mein substitute karo.

KYUN. Yahi poora point hai: exposure time position $x$ par depend karta hai, toh spread $x$ par depend karta hai. Plug karo aur simplify karo.

$δ \sim ν \cdot \frac{x}{U} = \frac{ν x}{U}$

PICTURE. Finished sideways-parabola: $δ$ leading edge par $0$ se start hoke $x$ ki tarah curve up karta hai — kabhi straight ramp nahi.

Exact laminar (laminar) constant Blasius solution of the Navier–Stokes equations se $\sim$ ko $\approx$ mein sharpen karta hai: $δ (x) \approx \frac{5.0 x}{R e _{x}}$

Step 7 — Edge cases: kya picture extremes par bhi survive karti hai?

KYA. Formula ko wahan test karo jahan woh break ho sakta hai: leading edge, still fluid, inviscid fluid, aur bahut fast stream.

KYUN. Ek law jis par tum trust karte ho use har corner par sensible pictures deni chahiye, degenerate ones mein bhi — warna reader ek aisi wall se takraata hai jo derivation ne kabhi dikhaayi nahi.

PICTURE. Char mini-panels, har ek ek extreme, har ek formula ki prediction se matched.

Corner case	Formula kya kehta hai	Picture / meaning
Leading edge, $x = 0$	$δ = ν \cdot 0/ U = 0$	Layer zero thickness ke saath janam leti hai — fluid ke paas slow hone ka time hi nahi tha.
Inviscid fluid, $ν \to 0$	$δ \to 0$	Koi stickiness nahi, koi slowing spread nahi — koi boundary layer nahi. Exactly [[d'Alembert's paradox
Bahut fast stream, $U \to \infty$	$δ \to 0$	Parcels itni jaldi blow past hote hain ki diffusion climb nahi kar sakta — vanishingly thin layer.
Slow / still stream, $U \to 0$	$δ \to \infty$	Bahut zyada exposure time; "thin layer" idea break down hota hai — Prandtl ka split high $R e_{x}$ chahta hai.

Step 8 — Bonus picture: friction sabse zyada kahan hai

KYA. $δ$ ko wall stress $τ_{w} = μ \partial u / \partial y ∣_{y = 0} \sim μU / δ$ mein wapas feed karo.

KYUN. Wahi picture jo $δ$ deti hai, hume plate ke along skin-friction bhi bataati hai — free mein.

$τ_{w} \sim μ \frac{U}{δ} \sim μ \frac{U}{ν x / U} = μU \frac{U}{ν x} \propto \frac{1}{x}$

PICTURE. Wall-stress curve $1/ x$ ki tarah drop karti hai: leading edge par steepest velocity slope (thinnest $δ$ ) ka matlab friction wahan sabse bada hai aur downstream fade karta hai.

Ek picture mein poora summary

Sab kuch ek canvas par: plate, growing $x$ layer, velocity profile $0 \to U$ squeeze karta hua, diffusion arrow $ν t$ ki tarah climb karta hua, aur exposure-time clock — woh chaar ideas jo $δ \sim ν x / U$ mein chain hote hain.

Recall Poore walkthrough ki Feynman retelling

Socho tum honey se dhaki table par haath ghisa rahe ho. Honey jo table ko touch kar rahi hai woh bilkul stuck hai — yahi no-slip hai (Step 1). Thoda upar woh thoda move karta hai, aur upar freely $U$ speed se flow karta hai; stuck se free tak ka woh squeeze ek thin band ki height $δ$ ke across hota hai (Step 2). Kyunki poora speed jump tiny height mein packed hai, honey layers ek doosre ke against hard shear kar rahi hain, aur woh rubbing force real hai chahe fluid thoda sticky ho — small stickiness times ek huge slope (Step 3). Ab "stuckness" ko kuch aisa socho jo upar spread karta hai ek dye blob ki tarah, ek rate par jo $ν$ se set hoti hai measured in area-per-time (Step 4). Ek honey parcel table ke kisi bhi patch ke upar sirf thode time $t = x / U$ ke liye hota hai, aur us time mein stuckness height $ν t$ tak climb karti hai (Step 5). Unhe saath rakho — $δ \sim ν x / U$ — aur tum ek aisi slow-zone paate ho jo downstream moti hoti jaati hai lekin dhire aur dhire, ek stretched sideways parabola ki tarah (Step 6). Bilkul shuru mein yeh zero-thick hai; stickiness khatam karo ya speed badha do aur yeh gayab ho jaati hai (Step 7); aur friction sabse zyada front par bites karta hai jahan layer sabse thin hai (Step 8). Yahi hai poori boundary layer, stuck honey se bani.